The integration of microfluidic channels, reservoirs, and control mechanisms onto a single chip is being applied to a wide variety of biomedical applications. So-called "lab-on-a-chip" (LOC) and micro-total-chemical-analysis-systems (u-TAS's) have already been developed for a number of standard laboratory procedures, including nucleotide sequencing, liquid chromatography, capillary electrophoresis and chemical synthesis. Due to their sensitivity and versatility, optical methods are generally used in these devices to probe for chemical reactions and separations via fluorescence and/or absorption. Unfortunately, available light sources are bulky and, in the case of UV lasers, very expensive. The development of a highly efficient, multi-element UV/visible light source which is commensuatewith the size and cost of the underlying microchip is proposed. Phase I will address key elements of risk associated with achieving the power denisty and spectral purity required to be compatible with important existing chemistry systems. Phase II will focus on integrating the results of Phase I into a light delivery sysem and demonstrating the utility of a plurality of individually progammable point sources. Because the proposed light source is separate and distinct from the chemistry layer, it can be readily customized for, and incorporated with, a wide variety of microfluidic devices. PROPOSED COMMERCIAL APPLICATION: The proposed commercial application is a compact, programmable light source for microfluidic devices/instruments.